Communications of the ACM
AI Powers Autonomous Materials Discovery
The Scientific Autonomous Reasoning Agent (SARA) integrates robotic materials synthesis using lateral gradient laser spike annealing and optical characterization, along with a hierarchy of artificial intelligence methods to map out processing phase diagra
Credit: Cornell University
Cornell University researchers have developed an artificial intelligence tool (AI) that aims to speed up the process of materials discovery.
In creating SARA (Scientific Autonomous Researching Agent for materials discovery and development), the researchers integrated robotic materials synthesis and characterization with a hierarchy of AI and active learning methods to uncover the structure of complex processing phase diagrams more quickly.
The researchers concentrated on inorganic materials that can be trapped in "metastable" states and potentially transformed to an "equilibrium" state over time.
SARA aims to reduce the amount of time and labor needed to identify metastable material from days to hours, and from hours to minutes.
Said Cornell's Michael Thompson, "The computer is controlling the experiment, in situ and live. There's a command to process material under particular conditions, and then immediately characterize it, and make a new decision about what the next experiment will be, based on the immediate new knowledge that's now available."
From Cornell University Chronicle
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found